Bacara jaoac vol 95 2012 508b


Published on

  • Be the first to comment

  • Be the first to like this

No Downloads
Total Views
On Slideshare
From Embeds
Number of Embeds
Embeds 0
No embeds

No notes for slide

Bacara jaoac vol 95 2012 508b

  1. 1. 446  TallenT eT al.: Journal of aoaC InTernaTIonal Vol. 95, no. 2, 2012 FOOD BIOLOGICAL CONTAMINANTS Efficient Isolation and Identification of Bacillus cereus Group Sandra M. TallenT U.S. Food and Drug Administration, Center for Food Safety and Applied Nutrition, 5100 Paint Branch Pkwy, College Park, MD 20740 KriSTin M. KoTewicz Oak Ridge Institute for Science and Education, MC-100-44, PO Box 117, Oak Ridge, TN 37831-0117 errol a. STrain and reginald w. BenneTT U.S. Food and Drug Administration, Center for Food Safety and Applied Nutrition, 5100 Paint Branch Pkwy, College Park, MD 20740 Bacillus cereus is a group of ubiquitous facultative anaerobic sporeforming Gram-positive rods commonly found in soil. The spores frequently contaminate a variety of foods, including produce, meat, eggs, and dairy products. Foodborne illnesses associated with toxins produced by B. cereus can result in self-limiting diarrhea or vomiting. Plate enumeration methods recommended by recognized food authorities to detect the presence of B. cereus in potentially contaminated food products do not inhibit other Gram-positive competitive bacteria. This study evaluated the use of Bacara, a new chromogenic agar, as an efficient method to identify and enumerate B. cereus group from food matrixes, even in the presence of background flora. Inclusivity and exclusivity testing was performed using four different selective and differential media for B. cereus, including Mannitol Egg Yolk Polymyxin (MYP), Polymyxin Pyruvate Egg-Yolk Mannitol Bromothymol Blue Agar, Bacillus Chromogenic Media, Brilliance, and Bacara. MYP and Bacara were also used in plate enumeration studies to isolate B. cereus from artificially contaminated foods. B acillus cereus has been detected and implicated in several contaminated food products and supplements since 1906, when Plazikowski associated the organism associated with food poisoning (1). The current method recommended in the U.S. Food and Drug Administration’s Bacteriological Analytical Manual (BAM; 2), International Organization for Standardization (ISO; 3, 4), and AOAC Official Methods of Analysis (5) for the enumeration and identification of B. cereus includes growth on Mannitol Egg Yolk Polymyxin (MYP) media and biochemical characterization of suspect isolates. Problematic issues with MYP, as noted by other authors, include a lack of characteristic colony morphology and production of lecithinase. Characteristic growth may be masked by the presence of background flora, such as Bacillus species other than B. cereus and Staphylococcus aureus, that ferment mannitol and produce lecithinase (6). Overlapping precipitation zones and the propensity of B. cereus colonies to coalesce hinders accurate enumeration (7). Lastly, the media must be used within 4 days of preparation (6, 7). Newer chromogenic media formulations incorporate specific Received June 8, 2011. Accepted by AH September 1, 2011. Corresponding author’s e-mail: DOI: 10.5740/jaoacint.11-251 enzyme substrates to allow for the identification of several organisms, including Escherichia coli, S. aureus, and Listeria monocytogenes. The use of selective media can eliminate the need for subcultures and biochemical testing resulting in more rapid detection and reduced reagent costs (8,9). Several of the formulations comply with food and water regulatory authorities (8). Previous B. cereus enumeration surveys using isolate panels, as well as food matrixes, have demonstrated the advantage of chromogenic media over traditional plating media, such as MYP or Polymixin Pyruvate Egg-Yolk Mannitol Bromothymol Blue Agar (PEMBA). The chromogenic agar tested in two separate studies clearly illustrated that the chromogenic media were more selective (7,10). Statistical information from one study comparing 100 B. cereus strains on chromogenic media confirmed that only three B. cereus strains showed weak or atypical reactions using the chromogenic media, compared to 75 B. cereus strains with weak or atypical reactions on PEMBA and 65 with weak or atypical reactions on MYP (10). The goal of this study was to compare the efficiency and accuracy of B.cereus enumeration from adulterated food matrixes using MYP and Bacara, a new chromogenic agar from AES Chemunex (Cranberry, NJ). Materials and Methods Bacterial Strains and Media A variety of selective and/or differential media were evaluated using previously characterized bacterial isolates obtained from American Type Culture Collection (Manassas, VA), Bacillus Genetic Stock Center (Columbus, OH), and U.S. Food and Drug Administration (FDA) collections (College Park, MD). The FDA B.cereus strains were associated with foodborne outbreaks, but the implicated food sources are unknown. Isolates were retrieved from frozen glycerol stock cultures, streaked to tryptic soy agar (TSA), and incubated overnight at 30°C. A single colony of each strain was grown overnight in brain heart infusion broth (pH 7.4) with 1 g/L glucose (BHIG), and 10 µL aliquots of the overnight culture were streaked to solid media for isolation. A total of 107 bacterial strains—including 61 B. cereus group strains, 14 Bacillus sp. not cereus, 19 Gram- negative bacterial strains, and 13 Gram-positive bacterial strains—were streaked to selective and differential media (Tables 1 and 2).The media included dehydrated media purchased from Oxoid Ltd (Basingstoke, UK), prepared according to the manufacturer’s instructions, MYP, PEMBA, and BrillianceTM agar. Bacillus Chromogenic Media (BCM) was purchased from R&F Laboratories (Downers Grove, IL) and prepared according
  2. 2. TallenT eT al.: Journal of aoaC InTernaTIonal Vol. 95, no. 2, 2012 447 Table 1. Summary of Bacillus cereus isolates used in this study; toxin status also shown, if knowna B. cereus strain No. Diarrheal enterotoxin Emetic toxin PCR result Comment A11778 Pos Neg Standard media testing strain A14579 Pos Neg Environmental strain B6A1 Neg Neg Wild type isolate; original code T B6A2 Neg Neg Wild type isolate; original code T-HT B6A3 Pos Neg Wild type isolate; original code NRRL-569; strepto- mycin-resistant B6A4 Pos Neg Streptomycin-resistant mu- tant of NRRL-569 (B6A3) B6A8 Pos Neg Probiotic strain; orgignal code bactisubtyl B6A10 Pos Neg Wild type isolate; orginal code ATCC13472 and HER 1399 B6A15 Pos Neg Wild type isolate; original code ATCC 10987 and NRS 248 B6A17 Pos Neg Wild type isolate; original code ATCC 13472 B6A18 Pos Neg Wild type isolate; original code ATCC 15816 B6A25 Neg Neg Wild type; orginal code GGC-D16B B6G1 Pos Neg gerQA2 mutant; orginal code AM1311 B6G2 Pos Neg gerIA5 mutant; original code AM134 B6E1 Pos Neg Resistant to teracycline; original code GP7 B6G3 Pos Neg ala-1 mutant; original code AM1316 B6A9 Pos Neg Wild type isolate F3995A Unk Neg Food F3998A Unk Neg Food F4014A Unk Unk Food F4034A Unk Unk Food F4047A Unk Unk Food F4225A Pos Neg Food F4226A Pos Pos Food F4228A Pos Neg Food F4229A Pos Neg Food F4230A Pos Pos Food F13061 Unk Unk Food FM77 Pos Unk Food FTJL14b Pos Pos Food Table 1. (continued) B. cereus strain No. Diarrheal enterotoxin Emetic toxin PCR result Comment F196668 Pos Neg Food F180WPB Pos Neg Food F4227A Pos Pos Food F66 Unk Unk Food F60006 Pos Pos Food F1259 Unk Unk Food F1261 Unk Unk Food F1263 Unk Unk Food F1265 Unk Unk Food F1396 Unk Unk Food F3A Pos Neg Food F26 Pos Neg Food F96 Pos Neg Food F4A Pos Neg Food F6A Pos Neg Food F59 Pos Neg Food FTOL-12 Pos Neg Food a American Type Culture Collection (ATCC); B = Bacillus Genetic Stock Center (BGSC); F = U.S. Food and Drug Administration (FDA) collection; Pos = positive; Neg = nega- tive; Unk = unknown enterotoxin status; FDA strains were outbreak strains from unknown food sources. b Bennett et al. (ref. 12). to the manufacturer’s instructions. Bacara was purchased from AES Chemunex as prepared media. Enterotoxin Detection The Tecra Bacillus Diarrheoal Enterotoxin Visual Immunoassay (3M, St. Paul, MN) is an ELISAscreening kit used to detect NheA, a protein associated with the diarrheal symptoms of B. cereus. The ELISAkit was used to screen isolates following manufacturer’s guidelines for foods. Isolates were also assessed for the presence of the emetic toxin using a PCR assay (primers unpublished). Results as available are listed in Table 1. Colony Morphology The growth characteristics recorded included presence or absence of precipitation zone, if applicable, and colony morphology after overnight incubation at 30°C. The expected B. cereus colony when grown on MYP is pink, on PEMBA peacock blue, and on Bacara orange-pink. All three formulations include egg yolk. If the bacterium produces lecithinase, the colony is surrounded by a halo or precipitation zone. B. cereus colonies grown on Brilliance and BCM are turquoise green.
  3. 3. 448  TallenT eT al.: Journal of aoaC InTernaTIonal Vol. 95, no. 2, 2012 Table 2. Summary of bacterial strains not B. cereus and the growth characteristics noted following overnight incubation at 30°C a Strain No. Bacterial isolate MYP BACARA F872 Acinetobacter sp. NG NG F873 Acinetobacter sp. NG NG F825 Aeromonas hydrophila NG NG F176 Bacillus amyloliquifaciens Y/HALO NG F177 Bacillus amyloliquifaciens Y/HALO NG F178 Bacillus amyloliquifaciens NG NG F179 Bacillus amyloliquifaciens NG NG F180 Bacillus amyloliquifaciens NG NG F181 Bacillus amyloliquifaciens NG NG F182 Bacillus amyloliquifaciens NG NG F183 Bacillus amyloliquifaciens NG NG F_BA663 Bacillus anthrcis Sterne P/NO HALO O/NO HALO A61 Bacillus circulans Y/NO HALO NG FMYC Bacillus mycoides Y/HALO NG F1267 Bacillus subtilis NG NG F1268 Bacillus subtilis Y/HALO NG F1269 Bacillus subtilis Y/HALO NG F1270 Bacillus subtilis Y/HALO NG F1258 Bacillus thuringiensis P/HALO O/HALO F1260 Bacillus thuringiensis P/HALO O/HALO F1262 Bacillus thuringiensis P/HALO O/HALO F1264 Bacillus thuringiensis P/HALO O/HALO F1266 Bacillus thuringiensis P/HALO O/HALO F1397 Bacillus thuringiensis P/HALO O/HALO F1398 Bacillus thuringiensis P/HALO O/HALO B6A21 Bacillus weihenstephanensis P/HALO O/HALO B6A22 Bacillus weihenstephanensis P/HALO O/HALO B6A23 Bacillus weihenstephanensis P/HALO O/HALO A64 Brevibacillus laterosporus P/NO HALO NG A25408 Citrobacter freundii NG NG A43864 Citrobacter koseri NG NG A13048 Enterobacter aerogenes NG NG A33731 Enterobacter amnigenus NG NG A35956 Enterobacter asburiae NG NG A35030 Enterobacter cloaceae NG NG F611 Enterobacter sakazakii NG NG A29212 Enterococcus faecalis Y/NO HALO O/NO HALO FNES23 Enterococcus faecalis Y/NO HALO O/NO HALO F385 Escherichia coli NG NG F386 Escherichia coli NG NG F387 Escherichia coli NG NG
  4. 4. TallenT eT al.: Journal of aoaC InTernaTIonal Vol. 95, no. 2, 2012 449 Table 2. (continued) Strain No. Bacterial Isolate MYP: 24 h BACARA: 24 h F392 Escherichia coli NG NG F1403 Listeria monocytogenes Y/HALO NG F1404 Listeria monocytogenes Y/HALO NG F1405 Listeria monocytogenes Y/HALO NG F1406 Listeria monocytogenes Y/HALO NG A29948 Paenibacillus gordonae NG NG A8509 Paenibacillus macerans Y/NO HALO NG A7070 Paenibacillus polymyxa Y/NO HALO NG F1 Salmonella enterica NG NG F249 Salmonella enterica NG NG F336 Salmonella enterica NG NG F522 Salmonella enterica NG NG F1161 Shigella sonnei NG NG A1766 Staph aureus Y/HALO W/NO HALO A1767 Staph aureus Y/HALO W/NO HALO A12228 Staph epidermidis Y/HALO NG A49051 Staph. intermedius P/HALO P/NO HALO A14576 Virgibacillus pantothenicus NG NG a A = American Type Culture Collection (ATCC); B = Bacillus Genetic Stock Center (BGSC); F = U.S. Food and Drug Administration (FDA) collection; NG = no growth; P = pink colony; Y = yellow colony; O = orange colony; W = white colony; HALO = lecithinase positive; NO HALO = lecithinase negative. Food Adulteration Food matrixes were artificially contaminated or adulterated with S. aureus (ATCC BAA1747), E. coli (FDA 392), Bacillus subtilis (FDA 1267), and an enterotoxingenic B. cereus strain (FDA 4227A) from a foodborne outbreak. The FDA 4227A strain tested positive for the diarrheal enterotoxin and was also presumptively positive for the emetic toxin by PCR screening. Single colonies of each strain were transferred to separate flasks with 50 mL BHIG and incubated overnight on an orbital shaker at 30°C. The B. cereus cells were pelleted and resuspended with 10 mL of Butterfield’s Phosphate Buffer (BPB), yielding approximately 1×10 9 CFU/mL. Low, medium, and high dilutions (approximately 1×10 –6 , 1×10 –4 , and 1×10 –1 CFU/mL, respectively) of B. cereus were prepared with BPB. A mixed culture was prepared using the resuspended pellets of S. aureus, E. coli, and B. subtilis. The pellets were combined in 20 mL BPB, and 6 mL of the triple species cocktail was combined with 6 mL of one of the B. cereus dilutions to spike a variety of food matrixes for enumeration studies, as described in the BAM (2). Briefly, 300 g food samples, including cooked rice, macaroni and cheese dinner, mashed potatoes, milk, dehydrated diet drink, and infant formula, were inoculated and mixed with bacterial cocktails and allowed to equilibrate at 4°C for 48 h. Five replicates of 50 g Table 3. Average colony counts (CFU/g) of five replicates plated in duplicate and transformed to log10 from low, medium, and high inoculum levels of B. cereus from artificially contaminated food samples MYP agar BACARA agar Sample Low Med High Low Med High Rice NDa ND ND 2.2 4.1 6.3 Diet drink ND ND 6.3 2.1 3.9 7.2 Infant formula ND ND ND 2.0 3.8 6.8 Mac and cheese ND ND ND 2.5 3.4 6.4 Mashed potatoes ND ND 5.9 2.8 4.1 6.8 Milk ND ND 7.6 2.3 4.5 7.6 a ND = No colonies detected.
  5. 5. 450  TallenT eT al.: Journal of aoaC InTernaTIonal Vol. 95, no. 2, 2012 Table 4. Results of a two-sample t-test on the log transformed data SD log10 CFU/g Product B. cereus target level BAM method, MYP New method, BACARA P value Rice 1 × 10–6 0.00 0.20 0.0044 Rice 1 × 10–4 0.00 0.56 0.0004 Rice 1 × 10 –1 0.00 1.19 0.0028 Mac and cheese 1 × 10 –6 0.00 0.30 0.2092 Mac and cheese 1 × 10–4 0.00 0.06 <0.0001 Mac and cheese 1 × 10–1 0.92 1.10 0.0841 Potato 1 × 10–6 0.00 0.38 0.0026 Potato 1 × 10 –4 0.00 0.26 <0.0001 Potato 1 × 10–1 0.77 1.08 0.0449 Milk 1 × 10 –6 0.00 0.30 0.0004 Milk 1 × 10 –4 0.00 0.32 <0.0001 Milk 1 × 10 –1 0.63 0.11 0.0005 Diet drink 1 × 10 –6 0.00 0.59 0.0433 Diet drink 1 × 10 –4 0.00 0.58 0.0008 Diet drink 1 × 10 –1 1.19 0.82 0.0096 Infant formula 1 × 10 –6 0.00 0.32 0.0264 Infant formula 1 × 10 –4 0.00 0.55 0.0006 Infant formula 1 × 10 –1 0.00 1.68 0.0093 portions of each food sample were homogenized at high speed (18 000–21 000 rpm) for 2 min in a Waring blender with 450 mL BPB. Ten-fold serial dilutions were prepared in BPB; 100 µL was then spread in duplicate to TSA, MYP, and Bacara for overnight incubation at 30°C. Due to a laboratory mishap, one set of plates was evaluated for the dehydrated diet drink. Plate Enumeration Analysis All aerobic plate counts inoculated from the six adulterated food matrixes were converted to CFU/g, and the average colony counts of five replicates plated in duplicate were transformed to base-10 logarithms (Table 3). The exception is that one set of plates was recorded for the diet drink. The log-transformed data were further analyzed (Table 4) by a two-sample t-test using the R version 2.11.1 software package (11). Results Figure 1. Colonies of B. cereus grown on MYP are pink and lecithinase positive (top left), but other bacteria are not inhibited and can obscure detection of B. cereus colonies (bottom left). Colonies of B. cereus grown on Bacara are pink-orange and are lecithinase positive (top right), but other organisms are inhibited (bottom right). Growth characteristics of B. cereus strains and B. cereus group strains showed typical colony morphology with a zone of precipitation if the media were supplemented with egg yolk. The comparison of BCM and Brilliance with the traditional media recommended by food authorities were in agreement with previous studies (7, 9, 10). However, Bacara was not included in the previous studies. Colonies of B. cereus and B. cereus group formed uniform individual pink-orange colonies on Bacara. Furthermore, the same strains with the exception of B. anthracis Sterne showed the typical zone of precipitation. The Gram-negative strains, including Acinetobacter sp., E. coli, and Enterotobacter sp., were inhibited on Bacara and MYP (Table 2). The Gram- positive bacterial strains, including Listeria monocytogenes, Paenibacillus sp., Virgibacillus sp., Bacillus amyloliquefaciens, B. circulans, B. mycoides, B. subtilis, Brevibacillus laterosporus, and Staphylococcus epidermidis, were also inhibited on Bacara, but some strains did grow on MYP (Table 2). Some of the staphylococci grew on Bacara as either white or pink pinpoint colonies; enterococci isolates grew on Bacara as pink pinpoint colonies. These colonies did not have a zone of precipitation and were easily distinguishable from representative B. cereus strains (Table 2). The bacterial growth on TSA plates from the adulterated food samples inoculated with B. cereus, B. subtilis, S. aureus, and E. coli was not much different than the growth on MYP plates. The abundant growth of competitive bacteria present in the mixed cultures interfered with the growth, detection, and colony enumeration of B. cereus. However, Bacara inhibited the competitive flora, and colonies of B. cereus were easily enumerated (Figure 1). The lowest level of B. cereus recovered from a mixed culture on Bacara showed a log average count of 2, but the log average count was significantly higher on MYP plates, at 5.9 (Table 3). In fact, colonies of B. cereus were generally not detected on MYP plate cultures from the adulterated food due to the presence of competitive bacteria (Table 3). The results of the statistical analysis showed a significant difference (P ≤0.05) in log CFU/g between MYP and Bacara for
  6. 6. TallenT eT al.: Journal of aoaC InTernaTIonal Vol. 95, no. 2, 2012 451 rice, infant formula, mashed potatoes, and milk. Bacara is able to detect B. cereus at much lower concentrations than MYP in these food matrixes when they have been artificially contaminated with competitive bacteria and dilutions of the target organism. There was not a significant difference between the log CFU/g for Bacara and MYP for the detection of B. cereus in the presence of competitive bacteria from macaroni and cheese (Table 4). Discussion Interpretation of MYP plates of the adulterated food samples was complicated by the presence of competitive bacterial flora because detection of discrete mannitol-negative B. cereus colonies were masked by the mannitol positive colonies of S. aureus and B. subtilis. Even enumeration of pure cultures is difficult since the colonies coalesce and the precipitation zones overlap (6). The aerobic plate counts for MYP indicated that the agar was not selective or differential for the growth of B. cereus in the presence of competitive bacteria. However, the target organism was grown on the Bacara plates, despite the high numbers of competitive bacteria present in the adulterated food samples. Additionally, the enumeration of B. cereus was more efficient and accurate with Bacara than with MYP, because the colonies were discrete and uniform in color and size. The time to detection (≤24 h) without further subculturing or biochemical testing was consistent with other chromogenic media and permitted completion in significantly less time than traditional media (8, 9). As with other chromogenic media, Bacara does not require additional conformational testing and, therefore, would be considered beneficial as a cost- and time-saving measure (8, 9). The statistical analysis shows low P values for the food matrixes tested, indicating that the two media, MYP and Bacara, are not equivalent in the detection of B. cereus in the presence of competitive flora. Bacara has been validated by AFNOR (Saint-Davis, France; Certificate No. AES 10/10-0710) for the enumeration of strains of the B. cereus group from food matrixes. We concur with the AFNOR laboratories and suggest that Bacara replace MYP as a more rapid method to recover B. cereus from contaminated food products. References